Forecasting the development trends of random processes using ordinal statistics
Abstract
Forecasting the development trends of random processes using ordinal statistics
Incoming article date: 28.11.2018The article presents the results of the development of a method for predicting trends in the development of random processes based on ordinal statistics, namely, the median and Hodges-Lehman statistics. The article discusses the proposed method in detail: it provides basic definitions, formulas for the calculation, a detailed description of the algorithm. The algorithm is implemented in the form of a software module that has practical application and can be used to solve problems of forecasting the trend of financial time series. The comparative results of applying the prediction method for the RTS index in the case of the use of the median and the Hodges-Lehmann statistics are given.
Keywords: ordinal statistics, random variable, random process, median, Hodges-Lehman statistics, forecasting, trend, mean square error, mean absolute error